ABSTRACT The cellular architecture and connectivity of the vertebrate retina is remarkably conserved across species. What distinguishes these retinas most is the relative numbers of each of the different cell types. Even within a species, there is significant variation in the size of neuronal populations. Polymorphic genes controlling the processes regulating cellular production, fate assignment and survival contribute most of this variation between individuals, and the present investigation will seek to identify genes responsible for this variation, and to understand their role in those processes modulating proliferation, fate determination and apoptosis. Twenty-six recombinant inbred strains of mice derived from the A/J and C57BL/6J strains of mice will be used to determine the natural variation in four different types of retinal bipolar cell, and the degree of co-variation between cell types will be examined. The variation in bipolar cell number will also be compared with data for cone photoreceptor number and with new data collected for rod photoreceptors. Variation in cell number will be mapped to genomic loci where candidate polymorphic genes will be identified and tested using gene knock-out strategies. The role of cell death in establishing bipolar cell numbers, and its temporal occurrence, will be assessed directly in Bax knock-out mice, while its afferent-dependency will be defined in coneless and conefull mutant mice. Other cell types have been shown to have their morphological differentiation controlled by the density of neighboring like-type cells as well as by their afferents, and so each of these variables will be modulated to determine their effects upon the differentiation of bipolar cell dendrites. Finally, a developmental transcriptome analysis of the retina will be conducted in these each of these recombinant inbred strains, and made available to the scientific community on-line at NerveNetwork. This will enable the direct mapping of variations in gene expression to genomic loci, thereby aiding in the identification of candidate genes underlying the above variations in bipolar and photoreceptor cell number, and the detection of correlations in gene expression to identify regulatory networks that participate in the production of individual types of bipolar or photoreceptor cells. These experiments will reveal the determinants of nerve cell number and morphology, clarifying our understanding of retinal development, as well as identifying gene polymorphisms that may contribute to retinal disease.